The present study examines how changes in the amount of on-screen text will influence student learning from a multimedia instructional unit on basic concepts of coordinate geometry. The relative effectiveness of two different versions (short-text and whole-text) of the instructional unit was examined for students who differed in terms of their ability to remember symbolic units, symbolic systems and symbolic interpretations. A total of 101 seventh graders were randomly assigned to work with either the whole-text or the short-text version. Student gains were analyzed using pre-test, post-test and retention test scores. Memory ability was assessed by the sub-tests of the Structure of Intellect-Learning Abilities Test. Results indicated no significant differences between groups who worked with short-text and whole-text versions. However retention scores of high and low memory groups who worked with the whole-text version showed significant differences. The whole-text version was observed to favor students with high memory for symbolic implications. Results suggest that workability of design principles for multimedia instruction may depend on the nature of the task and characteristics of the learner.